APPLICANT'S ABSTRACT: Recent clinical reports indicate a higher than normal incidence of hemorrhagic stroke, brain edema, and aneurysmal subarachnoid hemorrhage among light to heavy users of alcohol and a higher than normal incidence of aneurysmal subarachnoid hemorrhage, brain edema, and occlusion-type (non-hemorrhagic) stroke among users of cocaine. Some substances of abuse, particularly alcohol, are known to produce a loss in brain Mg. In the rat brain, alcohol and cocaine can produce spasm of microscopic arterioles and venules, followed by rupture of postcapillary venules and collecting venules, with frank bleeding into the perivascular space between the cortex and dura matter. Preliminary, in-situ experiments indicate that administration of Mg salts results in dose-dependent inhibition of the alcohol and cocaine-induced cerebral microvascular spasms and prevention and attenuation of the venular vasculotoxicity and hemorrhaging. A simple reduction in the CSF Mg2+ level, bathing the cortical microvessels, results in similar spasms, vasculotoxicity and hemorrhaging. In addition, we have demonstrated that chronic alcohol treatment depletes vascular smooth muscle cells of Mg and concomitantly results in Ca loading. The broad goal of the proposed research is to unravel the mechanism whereby alcohol and cocaine induce stroke-like events in the brain. The specific aims of the proposal, using biophysical, non-invasive techniques, as well as direct, quantitative in situ microscopy and radioisotopes are as follows: (1) To test the hypothesis that dietary manipulation in Mg intake will alter the responsiveness of in-situ cerebral microvessels to administration of vasculotoxic doses of alcohol and cocaine HC1. (2) Test the hypothesis that alcohol and cocaine-induced cerebrovasospasm and stroke-like events in the rat brain occur by virtue of loss of cellular Mg2+ coupled with disturbances in cellular bioenergetics. 31Phosphorus-nuclear magnetic resonance spectroscopy (31P-NMRS) in-vivo will be used to assess intracellular free Mg2+ ([Mg2+]i), cytosolic creatine phosphate (CrP), cytosolic adenosine triphosphate (ATP), cytosolic inorganic phosphate (Pi) and intracellular pH ([H+)i). Surface reflectance spectroscopy (SRS) will assess the mitochondrial level of reduced cytochrome oxidase aa3. (3) Test the hypothesis that (a) low dietary intake of Mg results in loss of brain Mg content concomitant with elevation in brain Ca content, and (b) alcohol or cocaine exacerbates these changes in divalent metal contents. (4) Using perfused rat brain slices (hippocampal and neocortex) and isolated canine basilar and middle cerebral arteries, 31P-NMRS, and 19F-NMRS and radiolabelled Ca, test the hypothesis that in-vitro addition of alcohol and cocaine result in concentration-dependent deficits in intracellular and cytosolic levels of Mg2+, CrP and ATP as well as elevation in [CA2+]i Pi and [H+]i. Using perfused rat brain slices and canine cerebral arteries and SRS, test the hypothesis that the mitochondrial level of reduced cytochrome oxidase aa3 is increased by alcohol and cocaine and ameliorated by Mg2+. (5) Test the hypothesis that alcohol and cocaine cause cerebrovasospasm by acting upon excitatory amino acid receptors (e.g., N-methyl-D-aspartate (NMDA) type of receptor), which are regulated by Mg2+. Mg2+ is known to block both voltage-sensitive, receptor-operated and NMDA-activated channels that allow CA2+ into neurons. NMDA receptor antagonists should therefore attenuate or ameliorate the ability of alcohol and cocaine to cause cerebrovasospasm both in-situ in the brain and in-vitro on isolated cerebral arteries.